Processor-based demultiplexing engines may be required to maintain flexibility while processing incoming data at a high rate. Ideally, the size of instruction memory for such a system is kept as small as possible. If full firmware parsing of each individual format is implemented, the firmware would have a separate path established for each part of the parsing tree, and as a result the code size and processing time become very large. For example, to parse one type of data format versus another type, the parsing and error checking are sometimes completely different between the different formats. This results in more code, and longer processing time.
Further limitations and disadvantages of conventional and traditional approaches will become apparent to one of skill in the art, through comparison of such systems with some aspects of the present invention as set forth in the remainder of the present application with reference to the drawings.